A Nomogram for Predicting Overall Survival in Primary Central Nervous System Lymphoma: A Retrospective Study.

in Journal of inflammation research by Yunan Ling, Xiaqi Miao, Xiang Zhou, Jingjing Ma, Zhiguang Lin, Qing Li, Mengxue Zhang, Yan Ma, Bobin Chen

TLDR

  • A novel prognostic model was developed to predict overall survival (OS) in patients with PCNSL, demonstrating high concordance Index and superior risk stratification compared to existing scoring systems.
  • The model provides valuable clinical guidance for decision making and can be used to accurately predict patient outcomes.

Abstract

Current prognostic scoring systems for newly diagnosed primary central nervous system lymphoma (PCNSL), such as IELSG prognostic score and MSKCC prognostic score, are widely used but have limitations in clinical practice. This study aimed to develop a novel prognostic model based on real clinical data and compare it with existing systems. A total of 288 patients newly diagnosed with PCNSL were recruited. Patients were randomly allocated to the development and validation cohorts. The least absolute shrinkage and selection operator (LASSO) regression and multivariate Cox regression analysis were used to identify the risk factors for overall survival (OS) and construct a nomogram. Additionally, Kaplan-Meier survival curves were plotted to show the stratification ability of the risk groups. Eastern Cooperative Oncology Group performance status (ECOG-PS), albumin, and two inflammatory biomarkers D-Dimer, and neutrophil-to-lymphocyte ratio (NLR)-were independent predictors of inferior OS. The prognostic model demonstrated concordance Index (C-index) of 0.731 and 0.679 in the development and validation cohorts, respectively. In terms of the time dependent area under the curve (AUC) values for OS, the development cohort exhibited values of 0.765, 0.762, and 0.812 for 1-year, 3-year, and 5-year OS, respectively. The corresponding AUC values in the validation cohort were 0.711, 0.731, and 0.840, respectively. The calibration curves showed excellent concordance. The novel prognostic model also provided superior risk stratification for patients with PCNSL compared with existing scoring systems. This study presents a novel prognostic model for predicting the OS of patients with newly diagnosed PCNSL. The model accurately and effectively stratifies the prognosis of patients with PCNSL and offers valuable clinical guidance for decision making.

Overview

  • The study aimed to develop a novel prognostic model for primary central nervous system lymphoma (PCNSL) based on real clinical data, comparing it with existing systems.
  • A total of 288 patients with newly diagnosed PCNSL were recruited and randomly allocated to the development and validation cohorts.
  • The primary objective is to develop a novel prognostic model that accurately predicts the overall survival (OS) of patients with PCNSL and offers valuable clinical guidance for decision making.

Comparative Analysis & Findings

  • The novel prognostic model demonstrated high concordance Index (C-index) of 0.731 and 0.679 in the development and validation cohorts, respectively.
  • The model showed superior risk stratification for patients with PCNSL compared to existing scoring systems, such as IELSG and MSKCC.
  • The calibration curves showed excellent concordance between the actual and predicted survival rates.

Implications and Future Directions

  • The novel prognostic model can be used to guide decision making in clinical practice, allowing for more accurate prediction of patient outcomes.
  • Future studies can focus on validating the model in larger and more diverse patient populations.
  • The model can be updated to include additional risk factors and biomarkers to further improve its performance.
Read Full Article